Bidirectional Generation of Structure and Properties Through a Single Molecular Foundation Model
This work addresses the problem of limited multimodal approaches in molecular AI for researchers and chemists, offering a versatile model for various chemical tasks, though it is incremental in building on existing multimodal learning techniques.
The authors tackled the lack of multimodal pre-training in molecular AI by developing a model that integrates structure and biochemical properties, enabling bidirectional information flow and achieving strong performance in tasks like conditional molecule generation and property prediction.
The recent success of large foundation models in artificial intelligence has prompted the emergence of chemical pre-trained models. Despite the growing interest in large molecular pre-trained models that provide informative representations for downstream tasks, attempts for multimodal pre-training approaches on the molecule domain were limited. To address this, we present a novel multimodal molecular pre-trained model that incorporates the modalities of structure and biochemical properties, drawing inspiration from recent advances in multimodal learning techniques. Our proposed model pipeline of data handling and training objectives aligns the structure/property features in a common embedding space, which enables the model to regard bidirectional information between the molecules' structure and properties. These contributions emerge synergistic knowledge, allowing us to tackle both multimodal and unimodal downstream tasks through a single model. Through extensive experiments, we demonstrate that our model shows remarkable capabilities in solving various meaningful chemical challenges, including conditional molecule generation, property prediction, molecule classification, and reaction prediction.